{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:MRJW4FTHKOGOTMAZ3Z3NZ24LFU","short_pith_number":"pith:MRJW4FTH","schema_version":"1.0","canonical_sha256":"64536e1667538ce9b019de76dceb8b2d2e4d7c85b01b201436ef7775bea85e84","source":{"kind":"arxiv","id":"1904.11325","version":1},"attestation_state":"computed","paper":{"title":"Communication trade-offs for synchronized distributed SGD with large step size","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Aymeric Dieuleveut, Kumar Kshitij Patel","submitted_at":"2019-04-25T13:27:30Z","abstract_excerpt":"Synchronous mini-batch SGD is state-of-the-art for large-scale distributed machine learning. However, in practice, its convergence is bottlenecked by slow communication rounds between worker nodes. A natural solution to reduce communication is to use the \\emph{`local-SGD'} model in which the workers train their model independently and synchronize every once in a while. This algorithm improves the computation-communication trade-off but its convergence is not understood very well. We propose a non-asymptotic error analysis, which enables comparison to \\emph{one-shot averaging} i.e., a single co"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1904.11325","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-04-25T13:27:30Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"911f0ca34b8df0fd56ef584564c3fbfcc63fbf9dbd4421df70ee0ef02995472f","abstract_canon_sha256":"6f24c4729f848f882b05b5136e07f82b6411b0abbd10c1bf96c5b086f1473867"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:47:45.210069Z","signature_b64":"8TAVfOjlWJk5vzGlTJagzFpbWQA0hLFNGfEIoEV9h2XGbHf45yxZf8Uo4JfadV+d0daafVh/169Grt4vAzs7CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64536e1667538ce9b019de76dceb8b2d2e4d7c85b01b201436ef7775bea85e84","last_reissued_at":"2026-05-17T23:47:45.209508Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:47:45.209508Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Communication trade-offs for synchronized distributed SGD with large step size","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Aymeric Dieuleveut, Kumar Kshitij Patel","submitted_at":"2019-04-25T13:27:30Z","abstract_excerpt":"Synchronous mini-batch SGD is state-of-the-art for large-scale distributed machine learning. However, in practice, its convergence is bottlenecked by slow communication rounds between worker nodes. A natural solution to reduce communication is to use the \\emph{`local-SGD'} model in which the workers train their model independently and synchronize every once in a while. This algorithm improves the computation-communication trade-off but its convergence is not understood very well. We propose a non-asymptotic error analysis, which enables comparison to \\emph{one-shot averaging} i.e., a single co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.11325","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1904.11325","created_at":"2026-05-17T23:47:45.209582+00:00"},{"alias_kind":"arxiv_version","alias_value":"1904.11325v1","created_at":"2026-05-17T23:47:45.209582+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.11325","created_at":"2026-05-17T23:47:45.209582+00:00"},{"alias_kind":"pith_short_12","alias_value":"MRJW4FTHKOGO","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_16","alias_value":"MRJW4FTHKOGOTMAZ","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_8","alias_value":"MRJW4FTH","created_at":"2026-05-18T12:33:21.387695+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU","json":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU.json","graph_json":"https://pith.science/api/pith-number/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/graph.json","events_json":"https://pith.science/api/pith-number/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/events.json","paper":"https://pith.science/paper/MRJW4FTH"},"agent_actions":{"view_html":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU","download_json":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU.json","view_paper":"https://pith.science/paper/MRJW4FTH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1904.11325&json=true","fetch_graph":"https://pith.science/api/pith-number/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/graph.json","fetch_events":"https://pith.science/api/pith-number/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/action/storage_attestation","attest_author":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/action/author_attestation","sign_citation":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/action/citation_signature","submit_replication":"https://pith.science/pith/MRJW4FTHKOGOTMAZ3Z3NZ24LFU/action/replication_record"}},"created_at":"2026-05-17T23:47:45.209582+00:00","updated_at":"2026-05-17T23:47:45.209582+00:00"}